Technologies for Efficient Identity Recognition Based on Skin Features

1. One or more non-transitory computer-readable storage media comprising a plurality of instructions that, when executed by a compute device, cause the compute device to at least: identify colors of pixels in an image; filter the pixels based on the identified colors by: assessing if a red component of respective pixels and a green component of the respective pixels satisfy a first color range condition; assessing if the red component, the green component, and a blue component of the respective pixels satisfy a second color range condition; categorizing respective pixels that satisfy both the first color range condition and the second color range condition into a first group; and categorizing respective pixels that fail to satisfy at least one of the first color range condition or the second color range condition into a second group; identify pixels that are associated with skin of a person based on the filtered pixels; and determine one or more features of the skin based on the identified pixels, wherein the determined one or more features are to be used in a feature vector to determine an identity of the person.

2. The one or more non-transitory computer-readable storage media of claim 1 , wherein to determine one or more features of the skin based on the identified pixels, the instructions cause the compute device to determine the one or more features of the skin based on a color of respective ones of the identified pixels.

3. The one or more non-transitory computer-readable storage media of claim 1 , wherein to determine the one or more features, the instructions cause the compute device to: determine a mode for respective ones of red, green, blue, and gray values of the identified pixels; and determine an average of respective ones of the red, green, and blue values.

4. The one or more non-transitory computer-readable storage media of claim 3 , wherein to determine the one or more features, the instructions cause the compute device to determine normalized values of the averages of the red, green, and blue values.

5. The one or more non-transitory computer-readable storage media of claim 1 , wherein to analyze the feature vector, the instructions cause the compute device to provide the feature vector to a support vector machine (SVM) classifier with a pre-trained SVM model that is based on other images that depict skin.

6. The one or more non-transitory computer-readable storage media of claim 1 , wherein to identify pixels that are associated with the skin of the person, the instructions cause the compute device to pixels that are associated with the skin of the person in an image of the face of the person.

7. The one or more non-transitory computer-readable storage media of claim 1 , wherein to identify pixels that are associated with the skin of the person, the instructions cause the compute device to identify pixels that are associated with the skin of the person in an image of a palm of the person.

8. A method comprising: identifying, by a compute device, colors of pixels in an image; filtering the pixels based on the identified colors by: assessing if a red component of respective pixels and a green component of the respective pixels satisfy a first color range condition; assessing if the red component, the green component, and a blue component of the respective pixels satisfy a second color range condition; categorizing respective pixels that satisfy both the first color range condition and the second color range condition into a first group; and categorizing respective pixels that fail to satisfy at least one of the first color range condition or the second color range condition into a second group; identifying pixels that are associated with skin of a person based on the filtered pixels; and determining, by the compute device, one or more features of the skin based on the identified pixels, wherein the determined one or more features are to be used in a feature vector to determine an identity of the person.

9. The method of claim 8 , wherein the identifying of the pixels in the image that are associated with the skin includes setting, to a predefined value, component color values of the pixels in the image that not associated with skin.

10. A compute device comprising: memory; and circuitry to: identify, in an image, colors of pixels; filter the pixels based on the identified colors by: assessing if a red component of respective pixels and a green component of the respective pixels satisfy a first color range condition; assessing if the red component, the green component, and a blue component of the respective pixels satisfy a second color range condition; categorizing respective pixels that satisfy both the first color range condition and the second color range condition into a first group; and categorizing respective pixels that fail to satisfy at least one of the first color range condition or the second color range condition into a second group; identify the pixels in the first group as associated with skin of a person; and determine one or more features of the skin based on the identified pixels, wherein the determined one or more features are to be used in a feature vector to determine an identity of the person.

11. The compute device of claim 10 , wherein to determine the one or more features, the circuitry is to: determine a mode for respective ones of red, green, blue, and gray values of the identified pixels; and determine an average of respective ones of the red, green, and blue values.

12. The compute device of claim 11 , wherein to determine the one or more features, the circuitry is to determine normalized values of the averages of the red, green, and blue values.

13. The compute device of claim 10 , wherein the circuitry is to provide the feature vector to a support vector machine (SVM) classifier with a pre-trained SVM model that is based on other images that depict skin.

14. The compute device of claim 10 , wherein to identify pixels that are associated with the skin of the person, the circuitry is to identify pixels that are associated with the skin of the person in an image of the face of the person.

15. The compute device of claim 10 , wherein to identify pixels that are associated with the skin of the person, the circuitry is to identify pixels that are associated with the skin of the person in an image of a palm of the person.

16. The compute device of claim 10 , wherein the circuitry is to output an indication of the determined identity and a confidence value indicative of a confidence that the determined identity is accurate.

17. The compute device of claim 10 , wherein the circuitry is to determine a distance between the eyes of the person based on the image and add the determined distance to the feature vector.

18. The compute device of claim 10 , wherein the circuitry is to set one or more color values of the pixels in the second group to a value indicative that the corresponding pixel is not representative of skin.

19. A compute device comprising: memory; and circuitry to: identify, in an image, colors of pixels; filter the pixels based on the identified colors; identify pixels that are associated with skin of a person based on the filtered pixels; determine one or more features of the skin based on the identified pixels, wherein the determined one or more features are to be used in a feature vector to determine an identity of the person; and determine a distance between the eyes of the person based on the image and add the determined distance to the feature vector, to determine the distance between the eyes of the person, the circuitry is to: convert the image from color to a grayscale image; generate a binary black and white image based on an auto-thresholding of the grayscale image; identify edges in the binary black and white image; crop the binary black and white image to an area that depicts eyes of the person based on the identified edges; identify blocks within the binary black and white image that contain the eyes; and determine, as the distance between the eye, a Euclidean distance between centers of the blocks.

20. The compute device of claim 19 , wherein to identify the pixels associated with the skin, the circuitry is to identify pixels that have a color component that satisfies a predefined relationship with other color components of the same pixel.

21. The compute device of claim 19 , wherein to determine the one or more features of the skin based on the identified pixels, the circuitry is to determine the one or more features of the skin based on a color of respective ones of the identified pixels.

22. One or more non-transitory computer-readable storage media comprising a plurality of instructions that, when executed by a compute device, cause the compute device to at least: identify, in an image, colors of pixels; filter the pixels based on the identified colors; identify pixels that are associated with skin of a person based on the filtered pixels; determine one or more features of the skin based on the identified pixels, wherein the determined one or more features are to be used in a feature vector to determine an identity of the person; and determine a distance between the eyes of the person based on the image and add the determined distance to the feature vector by: converting the image from color to a grayscale image; generating a binary black and white image based on an auto-thresholding of the grayscale image; identifying edges in the binary black and white image; cropping the binary black and white image to an area that depicts eyes of the person based on the identified edges; identifying blocks within the binary black and white image that contain the eyes; and determining, as the distance between the eye, a Euclidean distance between centers of the blocks.

23. The one or more non-transitory computer-readable storage media of claim 22 wherein to identify pixels associated with the skin, the instructions cause the compute device to identify pixels that have a color component that satisfies a predefined relationship with other color components of the same pixel.